Group B Streptococci (GBS) are a major cause of serious neonatal bacterial infections in this country. The origin of the bacteria for babies born with early-onset GBS infections is the birth canal of their mothers. This application describes an entirely new approach for preventing or eliminating vaginal carriage of GBS, thereby reducing the incidence of neonatal GBS infections. The goal is to genetically engineer a normal commensal organism present in the vagina, a Lactobacillus, to secrete a substance that will specifically prevent the growth of GBS. The substance selected is a GBS phage lysin that degrades the cell walls of GBS. The first specific aim is to clone and express the GBS phage lysin in E. coli and assess the bacteriocidal activity of purified recombinant enzyme. The ability of the purified lysin to kill GBS and various other bacteria will be first studied in vitro. Then it will then be determined if vaginal instillation of the enzyme will eliminate GBS in vaginally colonized mice. The second specific aim is to determine basic biochemical properties of the GBS phage lysin, including identifying the enzyme class to which it belongs and the identity of the cell wall component(s) necessary for specific peptidoglycan cleavage. The third specific aim is to engineer a Lactobacillus to secrete the GBS phage lysin. This will be done initially using a plasmid construct containing all the elements necessary for efficient secretion of the lysin from the transformed Lactobacillus. Then, in order to overcome inherent plasmid instability and also to eliminate the antibiotic selection markers, the lysin gene secretion cassette will be integrated into the chromosome of the Lactobacillus. The final specific aim is to determine if vaginal colonization of mice with the engineered Lactobacillus will prevent GBS from establishing a persistent colonization. In addition, it will also be determined if vaginal inoculation with the lactobacillus engineered to secrete GBS phage lysin will result in the clearance of GBS from the vaginas of mice previously colonized with the organism. The proposed research may lead to the development of an effective new method for long-term inhibition of GBS vaginal colonization, even though women may be repeatedly re-exposed to the bacteria. The method is unlikely to disturb the normal bacterial flora and should be very safe. If successful, this approach might also be used to protect against other genital, oral, and intestinal pathogens, especially those to which no effective mucosal immunity appears to develop.